Frontal Face Synthesis from Low-Resolution Images

A computer-implemented method for frontal face synthesis, the method comprising:

• synthesizing a high-resolution (HR) frontal-view (FV) image of a face from received features of a low-resolution (LR) non-frontal-view (NFV) image of the face, the received features input to a main path and an auxiliary path split from the main path, the HR FV image being of a higher resolution relative to a lower resolution of the LR NFV image,
• the synthesizing including producing, in the auxiliary path, auxiliary-path features from the received features and feeding the auxiliary-path features produced from the auxiliary path into the main path for synthesizing the HR FV image, the auxiliary-path features representing a HR NFV image of the face at the higher resolution.

A non-transitory computer-readable medium for frontal face synthesis, the non-transitory computer-readable medium having encoded thereon a sequence of instructions which, when loaded and executed by a processor, causes the processor to:

• synthesize a high-resolution (HR) frontal-view (FV) image of a face from received features of a low-resolution (LR) non-frontal-view (NFV) image of the face, the received features input to a main path and an auxiliary path split from the main path, the HR FV image being of a higher resolution relative to a lower resolution of the LR NFV image,
• the sequence of instructions further configured to cause the processor to produce, in the auxiliary path, auxiliary-path features from the received features and feed the auxiliary-path features produced from the auxiliary path into the main path for synthesizing the HR FV image, the auxiliary-path features representing a HR NFV image of the face at the higher resolution.